Just Shop It!

An excerpt from another message from my favorite organization for sending e-mails supporting policies I disagree with:

We're outraged that in the last year health insurance companies earned 56% more in profits and covered 2.7 million fewer people. And, healthcare insurance rate hikes of up to 39 percent are now happening across the country!

Why don't they stop being outraged for a moment and shop their policies? If their current vendor is such a rip-off, it sounds like a great opportunity for another insurance vendor to get its foot in the door. I cannot understand the mindset of just sitting and stewing about what your vendors are doing. It seems simple just to shop the quote to keep them honest and spend most of your effort finding ways to create value for your customers or employers.

I say this because I don't think the outrage is going to do anything. Even if the outrage make politicians carry out some very specific policy, it's not going to make people provide a service for free. Even if the government makes health insurance as we know it illegal, so the risk is sensibly spread among the entire population instead of being spread among those who bought insurance prior to be coming sick, we will still be paying the same amount overall for the same services.

What's "Good" for the Real Estate Market?

My broker’s newsletter today had an article called From Rescue to Recovery on the US housing market. USAA follows the same assumptions that all financial articles on real estate accept: Good = people moving house often and spending a lot of their money on housing. I suppose that is true from the real estate industry point of view, just as it’s may be good for refineries if people burn gasoline often and spend a lot of their money on gasoline. But we never hear this view in a financial article.

I believe in capitalism. I want people to create value that makes people spend their money on goods and services. If people aren’t buying your product, it feels bad for the person marketing the product, but it’s really just feedback from the market. Sometimes people don't want what you're selling.

Writers of articles on real estate should stop and ask themselves what are the virtues of people moving a lot and paying a lot for space.

It’s true that if a lot of good economic activity is going on in an area, it will make space there more expensive. But the cost of space is not a perfect barometer of the desirability of a particular area because we are coming out of a huge real estate bubble. It’s possible in the next few decades people will choose to spend more of their money on vacations instead of residential space. That doesn’t mean things are bad. It’s just changing preferences over time.

Crosstalk on a Reset Line Strikes Again

I have been working on an XScale board that has been giving me trouble for the month. Every time the software attempted to configure something on the mPCI slot, the system hung. Even though the problem only appeared after a spin of the board with the XScale processor, the software vendor (Team F1) worked assiduously on the problem refusing to blame the hardware.

It turns out the problem was a 1.3V core reset line going to the XScale processor was picking up crosstalk from the PCI bus because of a layout issue. The white line is the victim trace, which runs parallel to PCI lines. I do not have access to all artwork layers, but I think these traces run over a break in a split power plane.

The board designer didn’t think of a reset line as a critical trace because it doesn’t transition frequently. That keeps it from being an aggressor trace, but it’s very easy for a core reset line to be a victim of crosstalk. To be deasserted, this reset line must be greater than 1.0V. It doesn’t take much noise to pull it down into its indeterminate range (<1.0v) . I would have probably done the exact same thing because it's an easy mistake to make.

When I probed the line near the reset input, the 12pF capacitance of my scope probe was enough to prevent the problem. Tacking a 10pF cap to via going to that ball of the BGA made the problem go away altogether! Higher values worked just the same. At least this problem was easy to solve.

This is the second time in the past 14 months I’ve encountered a reset line being asserted due to noise or crosstalk. Reset lines must be treated as critical traces, especially if they’re low voltage and especially if their driven high only by weak pullup resistors.

The Basics of PCB Component Placement

There is a nice beginner's overview of component placement in the Feb 2010 edition of Printed Circuit Design and Fab magazing: Component Layout in Placement Process

Some interesting points:

• 75% of soldering defects are related to paste dispensing
• To deposit solderpaste on a large area it is better to use multiple smaller slots than one large slot.
• BGAs near a board’s edge are hard to rework because the thermal mass will be different on the side nearest the edge.

One thing missing from this overview: Even low-speed digital traces can be susceptible to noise from the outside world. It is not only important to keep clock lines clear of sensitive analog lines. It is important to keep digital lines, even “non-critical” ones with low edge rates and clock speeds, clear of places where noise from induced currents outside the board could corrupt the digital lines. A low-speed digital trace that veers across a mote and travels over a chassis ground plane is at risk of picking up noise from the outside world.

Rising Interest Rates Will Supress Real Estate Prices

My brokerage seems to agree with my notion that bond yields are poised to rise. Their bond fund manager says they're holding more cash than usual anticipating a drop in prices.

USAA: Have you taken some risk out of the portfolios you manage?

Espe: Yes. We think interest rates are going higher across the yield curve, and we've begun to prepare for it. Rates will probably go higher if the government stops buying Treasuries and mortgage-backed securities in March as planned. If they extend quantitative easing, it could be later, when the Fed raises rates as the economy recovers. But given the low current rates, the big budget deficit that has to be financed and the potential for the market to get spooked by any signs of inflation, we think higher rates are a question of when, not if.

The way we're managing risk is, first, by taking profits as bonds we own get back to normal valuations. Next, we have lowered exposure to rising rates by cutting back our duration, buying shorter-term securities and floating-rate securities. We're holding more cash than usual so that we are ready to deploy assets when rates rise. We have a big yield advantage over the peer group average, which will help in the event of defaults. We're also minimizing our holdings of U.S. Treasuries at these low interest rate levels, although we may increase our position if there's a spike in rates.

If interest rates rise, it's hard for me to see real estate prices increasing. I am baffled by people who have houses they want to sell but are holding on to them expecting real estate prices to increase.

Obama Should Support Manned Exploration of the Solar System

President Obama has proposed cancelling NASA’s Constellation program to send astronauts to the moon and to explore near-earth asteroids and Mars. He has proposed that once the space shuttle is retired NASA use private companies’ rockets to get astronauts into low-earth orbit. His proposal includes no plans to take humans beyond low-earth orbit.

President Obama is on the right track in creating partnerships with private companies for human space exploration. It is unfortunate, though, that he is calling for no provisions for human exploration of the solar system.

President Obama questions reusing old technologies from the 60s and 70s in a modern space program. I don’t know if reusing technologies is the right approach, but we certainly could use some of the enthusiasm for science and technology that we had in the 60s and 70s.

There are enormous intangible benefits to human exploration over probes. For example, my 18-month-old has recently gotten interested in the moon. This started with him seeing the moon in the sky, apparently independently of my interest in science and technology. I checked out some children’s books on the moon from the library for him. When he looks at the photographs and artist’s conceptions of the solar system, he focuses on “astronauts”. The Voyager probes taught us more about the solar system than human spaceflight has, but un-manned probes are not as exciting. We need people to be excited about science and technology.

We obviously need to be grownups and carry on using probes to do most of our space research. That visceral excitement of human exploration is more important that we might at first think.

I am glad Obama is looking into restructuring NASA and creating partnerships with private industry. When large numbers of humans being to work in space, for tourism, research, and industry, it will be private industry sponsoring it. None of this eliminates the benefits of a separate program for human exploration of the solar system.

LSR's Lab Helped Me Dig into an RF Issue

Yesterday I had the opportunity to spend the day in LSR’s lab in Cedarberg, WI working with their CTO, Brian Petted, on a physical layer issue in an 802.11 card. LSR’s has equipment that can pull out the modulation and display constellation diagram of digital signals and measure properties of the carrier.

My mission was to compare some 802.11b radios with performance issues to those that worked normally. Both radios met the 802.11b spec, but one had a much tighter constellation diagram than the other . (Yes, 802.11b is ancient, but this client has products with long industrial product life cycles.)

Good radio:

Bad radio:

The screenshots above contain loads of good info.
EVM[peak] – Average noise on the symbols, i.e. how much the constellation points spread out.
EVM[peak] – How far the worst outlier symbol lies from the median.
RMS Phase Noise – This is noise on the carrier itself.

It turned out the problem with the bad radio was that signal from the modulator was getting into the LO and causing phase noise in the carrier. Once there is phase noise on the carrier, you naturally have phase noise on the received symbols. Brian Petted worked out all of that without even removing the shield and probing the stages of the circuit.

It is very interesting that the 802.11b spec is so permissive. It allows huge symbol EVM, but compliant receivers do better if your transmitter has a nice tight constellation diagram.

Another interesting thing I found was that some radios I tested showed large side-lobes, approaching the limit mask, but worked fine in their application. Large side lobes seen on a spectrum analyzer can often indicate a problem, but sometimes they just mean a larger modulation index, which does not necessarily cause problems.

Usually when I’m testing wireless equipment, I have to work out what’s going on based on performance, usually PER vs SNR. I make hypotheses being careful not to tell them shape future observations; I can’t lose sight of the fact I don’t really know what’s going on at the physical layer. So it was amazing for me to spend some time in a lab with high-end equipment and a communications theory expert. I got insights that I never could have worked out sitting in my lab with a just scope, spectrum analyzer, and a WiFi sniffer.

I am always in awe when I dig into high-speed digital modulation because the technology completely blows away my wildest childhood geek-dreams I had back when I used a 2400bps landline modem and wireless digital modulation meant RTTY on shortwave or numeric pagers on VHF.